936 Transactions of tee American Institute. 



and as the Moors did not allow Christians to go to their universities, 

 he put on a turban and went to their famous universities at Cordova 

 and Seville as a Mahometan. When he went back to France he 

 carried with him the Arabic numerals ; and that was their first intro- 

 duction among the Christian nations of Europe. They were gradually 

 introduced, and without any special legislation they supplanted all the 

 systems of numeration which had been in use for a thousand years. 

 Gerbert may have been censurable for going there as he did; but it 

 is to be presumed that he repented, for he became archbishop of 

 Rheims, chancellor of France, and finally pope, under the name of 

 Sylvester II. To him we owe not only the Arabic notation, but a 

 demonstration of the fact that we can root out any usage, however 

 deeply entwined with the associations and habits of the people. 



But we have examples in our own time. When the British govern- 

 ment abolished their system of measures of capacity, there was no 

 difficulty about it ; and now even we Americans are occasionally 

 adopting the imperial gallon, although we have two gallons of our 

 own. Every year some of our countrymen go to Paris. It takes 

 them about ten days to learn the metric system of weights and meas- 

 ures, and then they have no further trouble. We have a much more 

 difficult system to learn, and yetjnen come to the United States by 

 thousands every year, and they change their habits without any com- 

 plaint, and soon talk about pounds and ounces as well as we do. 

 What we need is to familiarize the minds of the people with the 

 necessity for a change. 



The meter is founded upon a natural dimension, and therefore, if 

 by any accident the base should be lost, it would only be necessary to 

 measure an arc of the meridian again in order to restore it. It is 

 true it is asserted, as the result of modern English investigations, that 

 the meter is in error ; but how much do you think the error is ? It 

 is l-170th of an inch, a quantity so small that I doubt if there is a 

 person in this room who has a hair upon his head fine enough to indi- 

 cate that error. Let me tell you what I myself saw in 1867. The 

 arc of the meridian was measured in toiscs, and when the length of 

 the meter was ascertained, there were twelve standard meter bars 

 made of iron, and two of platinum, and four bars made equal to the 

 standard toise in length. These bars were verified, and the iron bars 

 sent to different countries as standards. One of them was brought to 

 this country by Mr. Hassler (afterward of the coast survey), in 1800. 

 Subsequently the French government sent over several other bars, 

 and upon a comparison of them there seemed to be a slight discrepancy. 



